Improvement of NIR prediction ability by dual model optimization in fusion of NSIA and SA methods

Li, Chunting; Chen, Huazhou; Zhang, Youyou; Hong, Shaoyong; Ai, Wu; Mo, Lina

doi:10.1016/j.saa.2022.121247

Cited by 14 publications

(5 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…After removing outliers from the HPLC dataset, the SPXY algorithm was used to divide 100 Epimedii Folium HPLC data into a training set (60 samples) and a validation set (40 samples) in a 3:2 ratio 20 . These 100 samples of Epimedii Folium were produced in Sichuan Province; 50 samples were collected in June and 50 samples were collected in August.…”

Section: Resultsmentioning

confidence: 99%

“…The NIR spectra were combined with the PLSR model to predict the absorbance values at each retention time point of the HPLC fingerprint to obtain the predicted HPLC fingerprint. Unlike quantitative models in which the contents of various indicators in Epimedii Folium are used as the predicted values (Y), in the fingerprint conversion process the absorbance values at each retention time point of the HPLC fingerprint are used as the predicted values (Y), while NIR spectral data are used as the independent variable (X).After removing outliers from the HPLC dataset, the SPXY algorithm was used to divide 100 Epimedii Folium HPLC data into a training set (60 samples) and a validation set (40 samples) in a 3:2 ratio 20. These 100 samples of Epimedii Folium were produced in Sichuan Province; 50 samples were collected in June and 50 samples were collected in August.Using the absorbance values of every retention time point in HPLC as the predicted values in the spectral preprocessing process is an ineffective and tedious task.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Holistic quality evaluation method of Epimedii Folium based on NIR spectroscopy and chemometrics

Chen,

Zhang,

Feng

et al. 2024

Phytochemical Analysis

View full text Add to dashboard Cite

IntroductionThere are some problems in the quality control of Epimedii Folium (leaves of Epimedium brevicornum Maxim.), such as the mixed use of Epimedii Folium from different harvesting periods and regions, incomplete quality evaluation, and time‐consuming analysis methods.ObjectiveNear‐infrared (NIR) spectroscopy was conducted to establish a rapid overall quality evaluation method for Epimedii Folium.Materials and methodsQuantitative models of the total solid, moisture, total flavonoid, and flavonol glycoside (Epimedin A, Epimedin B, Epimedin C, Icariin) contents of Epimedii Folium were established by partial least squares regression (PLSR). The root mean square error (RMSE) and correlation coefficient (R) were used to evaluate the performance of models. The qualitative models of Epimedii Folium from different geographic origins and harvest periods were established based on K‐nearest neighbor (KNN), back‐propagation neural network (BPNN), and random forest (RF). Accuracy and Kappa values were used to evaluate the performance of models. A new multivariable signal conversion strategy was proposed, which combines NIR spectroscopy with the PLSR model to predict the absorbance values of retention time points in the high‐performance liquid chromatography (HPLC) fingerprint to obtain the predicted HPLC fingerprint. The Pearson correlation coefficient and cosine coefficient were used to evaluate the similarity between real and predicted HPLC fingerprints.ResultsQualitative models, quantitative models, and the similarity between real and predicted HPLC fingerprints are satisfactory.ConclusionThe method serves as a fast and green analytical quality evaluation method of Epimedii Folium and can replace traditional methods to achieve the overall quality evaluation of Epimedii Folium.

show abstract

Section: Resultsmentioning

confidence: 99%

mentioning

confidence: 99%

Holistic quality evaluation method of Epimedii Folium based on NIR spectroscopy and chemometrics

Chen,

Zhang,

Feng

et al. 2024

Phytochemical Analysis

View full text Add to dashboard Cite

show abstract

“…2.0 ≤ RPD < 2.5 indicates that the model has a good predictive ability. And 2.5 ≤ RPD < 3.0 indicates that the model has very good predictive ability (Chen et al 2022, Li et al 2022a, 2023. Larger R 2 and RPD and smaller RMSE indicate that the model predicts better.…”

Section: Methodsmentioning

confidence: 99%

“…Prediction sets are then utilised to assess the predictive accuracy of the model. The sample set partitioning based on Joint X-Y Distance (SPXY) algorithm can optimise the search for samples and determine the most efficient way to partition them, with the objective of minimising the differences between the resulting subsets (Li et al 2022a). The SPXY algorithm was used to group all samples (N = 225) into two datasets, of which 169 (75% of the total) and 56 (25% of the total) samples formed the correction and prediction datasets, respectively.…”

Section: Statistical Analysis Of the Pattern Of Change Of The Ahnmentioning

confidence: 99%

Hyperspectral analysis of the content of the alkali-hydrolysed nitrogen in the soil of a millet field

Zhu,

Wang,

Zhang

et al. 2023

Plant Soil Environ.

View full text Add to dashboard Cite

Nitrogen, one of the key nutrients essential for plant growth in soil, plays a crucial role. Organic matter in the soil can be decomposed by microorganisms into inorganic nitrogen. The level of alkali-hydrolysed nitrogen (AHN) is increased, which is beneficial for plant absorption and utilisation. As a major source of nutrients for plant growth, soil nutrient content is an important indicator for evaluating the environment for plant growth. Rapid access to soil nutrient information is an important requirement for the development of modern precision agriculture. As a major component of soil, soil nitrogen is an important indicator of soil fertility. Under certain conditions, the nitrogen content level marks the soil fertility level. It is the main source of nutrients needed for plant growth. Soil nitrogen plays a critical role in plant growth and development. Nitrogen deficiency affects the protein content of crops. When organic matter decomposes, most of the nitrogen undergoes conversion to inorganic nitrogen and is then released into the soil (Tahmasbian et al. 2018).

show abstract

“…If the correction model is established directly with the wavelength variables (WVs) of the whole spectrum, the accuracy and robustness of the model will eventually be affected, due to the weak correlation between some spectral WVs and the components [ 13 ]. To effectively extract the characteristic WVs (CWVs) with high correlation and to establish a simpler and more stable NIRS model, scholars have proposed using interval partial least squares [ 14 ], synergy partial least squares [ 15 ], backward partial least squares (BIPLS) [ 16 ], and other spectral area optimization algorithms, together with uninformative variable elimination [ 17 ], competitive adaptive weighted sampling (CARS) [ 18 ], and various other wavelength selection algorithms, and genetic algorithms (GA) [ 19 ], genetic simulated annealing algorithms (GSA) [ 20 ], ant colony algorithms [ 21 ], particle swarm optimization algorithms [ 22 ], and various other intelligent optimization algorithms to effectively filter out WVs. Sometimes a single WV optimization method fails to meet the requirements of the analysis, and a combination of methods is required [ 23 , 24 , 25 ].…”

Section: Introductionmentioning

confidence: 99%

Rapid Determination of Cellulose and Hemicellulose Contents in Corn Stover Using Near-Infrared Spectroscopy Combined with Wavelength Selection

Wang

Feng

et al. 2022

Molecules

View full text Add to dashboard Cite

The contents of cellulose and hemicellulose (C and H) in corn stover (CS) have an important influence on its biochemical transformation and utilization. To rapidly detect the C and H contents in CS by near-infrared spectroscopy (NIRS), the characteristic wavelength selection algorithms of backward partial least squares (BIPLS), competitive adaptive reweighted sampling (CARS), BIPLS combined with CARS, BIPLS combined with a genetic simulated annealing algorithm (GSA), and CARS combined with a GSA were used to select the wavelength variables (WVs) for C and H, and the corresponding regression correction models were established. The results showed that five wavelength selection algorithms could effectively eliminate irrelevant redundant WVs, and their modeling performance was significantly superior to that of the full spectrum. Through comparison and analysis, it was found that CARS combined with GSA had the best comprehensive performance; the predictive root mean squared errors of the C and H regression model were 0.786% and 0.893%, and the residual predictive deviations were 3.815 and 12.435, respectively. The wavelength selection algorithm could effectively improve the accuracy of the quantitative analysis of C and H contents in CS by NIRS, providing theoretical support for the research and development of related online detection equipment.

show abstract

Improvement of NIR prediction ability by dual model optimization in fusion of NSIA and SA methods

Cited by 14 publications

References 41 publications

Holistic quality evaluation method of Epimedii Folium based on NIR spectroscopy and chemometrics

Holistic quality evaluation method of Epimedii Folium based on NIR spectroscopy and chemometrics

Hyperspectral analysis of the content of the alkali-hydrolysed nitrogen in the soil of a millet field

Rapid Determination of Cellulose and Hemicellulose Contents in Corn Stover Using Near-Infrared Spectroscopy Combined with Wavelength Selection

Contact Info

Product

Resources

About